Various inspection apparatus (hereinafter referred to as modality apparatus) used in image diagnosis can conduct inspections in the body in a low invasive manner and are indispensable in current medical practice. With improvement of performances of the modality apparatus, a good-quality image with high resolution can be now acquired, and accurate and precise inspections have become possible in image diagnosis. There are many imaging methods according to the modality apparatus such as an X-ray CT (Computed Tomography) apparatus capable of acquiring three-dimensional information of an internal tissue of an object with high resolution and an MRI (Magnetic Resonance Imaging) apparatus capable of MR Angiography (MRA) in which fresh blood without a contrast agent mixed is imaged, for example. Digitalization of medical images have progressed and an HIS (Hospital Information System) which is an ordering system for processing an inspection request from a doctor through an electronic network, an RIS (Radiology Information System), a PACS (Picture Archiving and Communication Systems) for storing images acquired by the modality apparatus as electronic data and the like have been developed.
As described above, development of the modality apparatus have enabled easy and detailed in-vivo observation. In contrast, an attainable data amount has become huge, and in many modality apparatus, data is acquired in a format of volume data constituted by a plurality of images. A data amount of the volume data amounts to several thousand shots when an image of a whole body is picked up, and a burden on a diagnostic reading doctor or the like by using such data so as to make diagnostic reading is large. Diagnostic reading is an important work in diagnosing a disease and determination of treatment policies and early detection is in demand, but analysis of a large amount of medical images and determination at an early stage is not easy. Thus, as an invention for supporting image diagnosis, a medical image processing apparatus for specifying an anatomical part by using a segmentation technology or the like and for determining an abnormal region and its malignancy and an image analyzing apparatus for determining a positional correspondence relation in images acquired in two different inspections by a structure having periodicity such as a spine are provided.
Diagnostic reading and diagnosis require accuracy, and to make high-quality diagnosis, abnormal parts and treatment parts need to be accurately grasped in the acquired medical images. However, diagnostic reading of anatomical parts from medical images requires skilled techniques and knowledge. Thus, technologies for expressing/constructing anatomical positions of a human body by using mathematical methods have been provided and studied.
The anatomical position refers to a position in a pathognomonic local structure (hereinafter referred to simply as a local structure) of a human body playing an important role in understanding the medical images and makes a mark when the human body is mapped anatomically. For example, an anterior arch (node) in a first cervical spine in a head, a bifurcation of trachea in a chest or an upper pole of right kidney in an abdomen corresponds to the local structure. A position (anatomical position) of this local structure is automatically detected from the medical images acquired by the modality apparatus such as an X-ray CT apparatus and an MRI apparatus by general image analysis, a pattern recognition technique and the like.
Diagnostic reading of medical images performed when creating a diagnostic reading report needs to be performed extensively so as not to miss finding. That is, all the slice images acquired in the number of several hundreds to thousands per patient in one session of imaging need to be checked. However, parts and local structures not relating to an inspection purpose are included in the slice images acquired in the inspection. Diagnostic reading doctors and the like need to analyze such large amount of medical image data, and whether or not parts and local structures which really need to be checked in diagnostic reading have been checked cannot be grasped easily. In diseases in which abnormality appears in a plurality of organs such as cancers and immune diseases, the relating parts or local structures might be missed by mistake such as mere lack of knowledge or forgetfulness.
Thus, a medical image processing apparatus capable of checking whether or not the parts and local structures requiring check in an inspection or diagnosis have been diagnostically read based on a position (anatomical position) of the above-described local structures is in demand.